arXiv subjects

Yongqiang Yang

Publications and source records attributed to Yongqiang Yang.

At least 19 recordsLinked to original sources

arXiv2026

Huawei Cloud Model-as-a-Service on the CloudMatrix384 SuperPod

Scaled-out MoE LLMs and scaled-up SuperPods create new systems challenges for production Model-as-a-Service (MaaS), requiring disaggregation, low-latency communication, and decentralized serving. This report presents xDeepServe, the production serving system behind Huawei Cloud's MaaS offering on CloudMatrix384, a 48-server SuperPod with 384 Ascend 910C chips connected by a high-bandwidth UB fabric and global shared memory. It serves models including DeepSeek, Kimi, GLM, Qwen, and MiniMax, among others. xDeepServe is built around Transformerless, a disaggregated execution architecture that decomposes transformer inference into modular units -- attention, feedforward, and MoE -- and supports disaggregated Prefill-Decode and MoE-Attention deployments. To enable disaggregation, we develop XCCL, a memory-semantic communication layer providing microsecond-level point-to-point and scalable all-to-all primitives, and we extend FlowServe with decentralized DP groups and techniques to mitigate stragglers and synchronization variance. In a peak decoding configuration, xDeepServe reaches 2400 tokens/s per Ascend 910C chip at ~50ms time-per-output-token (TPOT).

cs.DC↗

arXiv2025

Serving Large Language Models on Huawei CloudMatrix384

The rapid evolution of large language models (LLMs), driven by growing parameter scales, adoption of mixture-of-experts (MoE) architectures, and expanding context lengths, imposes unprecedented demands on AI infrastructure. Traditional AI clusters face limitations in compute intensity, memory bandwidth, inter-chip communication, and latency, compounded by variable workloads and strict service-level objectives. Addressing these issues requires fundamentally redesigned hardware-software integration. This paper introduces Huawei CloudMatrix, a next-generation AI datacenter architecture, realized in the production-grade CloudMatrix384 supernode. It integrates 384 Ascend 910 NPUs and 192 Kunpeng CPUs interconnected via an ultra-high-bandwidth Unified Bus (UB) network, enabling direct all-to-all communication and dynamic pooling of resources. These features optimize performance for communication-intensive operations, such as large-scale MoE expert parallelism and distributed key-value cache access. To fully leverage CloudMatrix384, we propose CloudMatrix-Infer, an advanced LLM serving solution incorporating three core innovations: a peer-to-peer serving architecture that independently scales prefill, decode, and caching; a large-scale expert parallelism strategy supporting EP320 via efficient UB-based token dispatch; and hardware-aware optimizations including specialized operators, microbatch-based pipelining, and INT8 quantization. Evaluation with the DeepSeek-R1 model shows CloudMatrix-Infer achieves state-of-the-art efficiency: prefill throughput of 6,688 tokens/s per NPU and decode throughput of 1,943 tokens/s per NPU (<50 ms TPOT). It effectively balances throughput and latency, sustaining 538 tokens/s per NPU even under stringent 15 ms latency constraints, while INT8 quantization maintains model accuracy across benchmarks.

cs.DC↗

arXiv2025

KPIRoot+: An Efficient Integrated Framework for Anomaly Detection and Root Cause Analysis in Large-Scale Cloud Systems

To ensure the reliability of cloud systems, their performance is monitored using KPIs (key performance indicators). When issues arise, root cause localization identifies KPIs responsible for service degradation, aiding in quick diagnosis and resolution. Traditional methods rely on similarity calculations, which can be ineffective in complex, interdependent cloud environments. While deep learning-based approaches model these dependencies better, they often face challenges such as high computational demands and lack of interpretability. To address these issues, KPIRoot is proposed as an efficient method combining similarity and causality analysis. It uses symbolic aggregate approximation for compact KPI representation, improving analysis efficiency. However, deployment in Cloud H revealed two drawbacks: 1) threshold-based anomaly detection misses some performance anomalies, and 2) SAX representation fails to capture intricate variation trends. KPIRoot+ addresses these limitations, outperforming eight state-of-the-art baselines by 2.9% to 35.7%, while reducing time cost by 34.7%. We also share our experience deploying KPIRoot in a large-scale cloud provider's production environment.

cs.SE↗

arXiv2025

LLMPrism: Black-box Performance Diagnosis for Production LLM Training Platforms

Large Language Models (LLMs) have brought about revolutionary changes in diverse fields, rendering LLM training of utmost importance for modern enterprises. To meet this demand, multi-tenant large-scale LLM training platforms have been built to offer LLM training services. Nevertheless, due to the complexity and synchronous nature of LLM training process, performance issues occur frequently and can result in substantial resource wastage. The limited visibility from the perspective of platform providers impedes existing profiling methods and poses challenges to the monitoring and diagnosis of the performance of LLM training jobs. For the first time, this paper proposes the utilization of underlying network flow data to reconstruct the training timelines of jobs based on the distinct characteristics in the LLM training procedure. We design LLMPrism, the first black-box performance diagnosis system for LLM training platforms. By progressively recognizing LLM training jobs, identifying their parallelism strategies, and reconstructing the training timelines, LLMPrism achieves non-intrusive, lightweight, and continuous monitoring of LLM training systems. Leveraging this monitoring capability, it further effectively diagnoses potential performance issues. Since Oct. 2024, LLMPrism has been deployed on our large-scale production Platform-X, in which the evaluations and deployment experiences demonstrate that LLMPrism can achieve accurate timeline reconstruction with an error within 0.3% and effectively diagnose various performance issues.

cs.SE↗

arXiv2025

L4: Diagnosing Large-scale LLM Training Failures via Automated Log Analysis

As Large Language Models (LLMs) show their capabilities across various applications, training customized LLMs has become essential for modern enterprises. However, due to the complexity of LLM training, which requires massive computational resources and extensive training time, failures are inevitable during the training process. These failures result in considerable waste of resource and time, highlighting the critical need for effective and efficient failure diagnosis to reduce the cost of LLM training. In this paper, we present the first empirical study on the failure reports of 428 LLM training failures in our production Platform-X between May 2023 and April 2024. Our study reveals that hardware and user faults are the predominant root causes, and current diagnosis processes rely heavily on training logs. Unfortunately, existing log-based diagnostic methods fall short in handling LLM training logs. Considering the unique features of LLM training, we identify three distinct patterns of LLM training logs: cross-job, spatial, and temporal patterns. We then introduce our Log-based Large-scale LLM training failure diagnosis framework, L4, which can automatically extract failure-indicating information (i.e., log events, nodes, stages, and iterations) from extensive training logs, thereby reducing manual effort and facilitating failure recovery. Experimental results on real-world datasets show that L4 outperforms existing approaches in identifying failure-indicating logs and localizing faulty nodes. Furthermore, L4 has been applied in Platform-X and demonstrated its effectiveness in enabling accurate and efficient failure diagnosis.

cs.SE↗

arXiv2025

Anomalous Meets Topological Hall Effect in Cr2Ge2Te6 Heterostructures

Introducing topologically protected skyrmions in graphene holds significant importance for developing high-speed, low-energy spintronic devices. Here, we present a centrosymmetric ferromagnetic graphene/trilayer Cr2Ge2Te6/graphene heterostructure, demonstrating the anomalous and topological Hall effect due to the magnetic proximity effect. Through gate voltage control, we effectively tune the emergence and size of skyrmions. Micromagnetic simulations reveal the formation of skyrmions and antiskyrmions, which respond differently to external magnetic fields, leading to oscillations in the topological Hall signal. Our findings provide a novel pathway for the formation and manipulation of skyrmions in centrosymmetric two-dimensional magnetic systems, offering significant insights for developing topological spintronics.

cond-mat.mes-hall↗

arXiv2024

Identifying Performance Issues in Cloud Service Systems Based on Relational-Temporal Features

Cloud systems are susceptible to performance issues, which may cause service-level agreement violations and financial losses. In current practice, crucial metrics are monitored periodically to provide insight into the operational status of components. Identifying performance issues is often formulated as an anomaly detection problem, which is tackled by analyzing each metric independently. However, this approach overlooks the complex dependencies existing among cloud components. Some graph neural network-based methods take both temporal and relational information into account, however, the correlation violations in the metrics that serve as indicators of underlying performance issues are difficult for them to identify. Furthermore, a large volume of components in a cloud system results in a vast array of noisy metrics. This complexity renders it impractical for engineers to fully comprehend the correlations, making it challenging to identify performance issues accurately. To address these limitations, we propose Identifying Performance Issues based on Relational-Temporal Features (ISOLATE ), a learning-based approach that leverages both the relational and temporal features of metrics to identify performance issues. In particular, it adopts a graph neural network with attention to characterizing the relations among metrics and extracts long-term and multi-scale temporal patterns using a GRU and a convolution network, respectively. The learned graph attention weights can be further used to localize the correlation-violated metrics. Moreover, to relieve the impact of noisy data, ISOLATE utilizes a positive unlabeled learning strategy that tags pseudo-labels based on a small portion of confirmed negative examples. Extensive evaluation on both public and industrial datasets shows that ISOLATE outperforms all baseline models with 0.945 F1-score and 0.920 Hit rate@3.

cs.LG↗

arXiv2024

MicroRes: Versatile Resilience Profiling in Microservices via Degradation Dissemination Indexing

Microservice resilience, the ability of microservices to recover from failures and continue providing reliable and responsive services, is crucial for cloud vendors. However, the current practice relies on manually configured rules specific to a certain microservice system, resulting in labor-intensity and flexibility issues, given the large scale and high dynamics of microservices. A more labor-efficient and versatile solution is desired. Our insight is that resilient deployment can effectively prevent the dissemination of degradation from system performance metrics to user-aware metrics, and the latter directly affects service quality. In other words, failures in a non-resilient deployment can impact both types of metrics, leading to user dissatisfaction. With this in mind, we propose MicroRes, the first versatile resilience profiling framework for microservices via degradation dissemination indexing. MicroRes first injects failures into microservices and collects available monitoring metrics. Then, it ranks the metrics according to their contributions to the overall service degradation. It produces a resilience index by how much the degradation is disseminated from system performance metrics to user-aware metrics. Higher degradation dissemination indicates lower resilience. We evaluate MicroRes on two open-source and one industrial microservice system. The experiments show MicroRes' efficient and effective resilience profiling of microservices. We also showcase MicroRes' practical usage in production.

cs.SE↗

arXiv2024

FaultProfIT: Hierarchical Fault Profiling of Incident Tickets in Large-scale Cloud Systems

Postmortem analysis is essential in the management of incidents within cloud systems, which provides valuable insights to improve system's reliability and robustness. At CloudA, fault pattern profiling is performed during the postmortem phase, which involves the classification of incidents' faults into unique categories, referred to as fault pattern. By aggregating and analyzing these fault patterns, engineers can discern common faults, vulnerable components and emerging fault trends. However, this process is currently conducted by manual labeling, which has inherent drawbacks. On the one hand, the sheer volume of incidents means only the most severe ones are analyzed, causing a skewed overview of fault patterns. On the other hand, the complexity of the task demands extensive domain knowledge, which leads to errors and inconsistencies. To address these limitations, we propose an automated approach, named FaultProfIT, for Fault pattern Profiling of Incident Tickets. It leverages hierarchy-guided contrastive learning to train a hierarchy-aware incident encoder and predicts fault patterns with enhanced incident representations. We evaluate FaultProfIT using the production incidents from CloudA. The results demonstrate that FaultProfIT outperforms state-of-the-art methods. Our ablation study and analysis also verify the effectiveness of hierarchy-guided contrastive learning. Additionally, we have deployed FaultProfIT at CloudA for six months. To date, FaultProfIT has analyzed 10,000+ incidents from 30+ cloud services, successfully revealing several fault trends that have informed system improvements.

cs.SE↗

arXiv2023

Prism: Revealing Hidden Functional Clusters from Massive Instances in Cloud Systems

Ensuring the reliability of cloud systems is critical for both cloud vendors and customers. Cloud systems often rely on virtualization techniques to create instances of hardware resources, such as virtual machines. However, virtualization hinders the observability of cloud systems, making it challenging to diagnose platform-level issues. To improve system observability, we propose to infer functional clusters of instances, i.e., groups of instances having similar functionalities. We first conduct a pilot study on a large-scale cloud system, i.e., Huawei Cloud, demonstrating that instances having similar functionalities share similar communication and resource usage patterns. Motivated by these findings, we formulate the identification of functional clusters as a clustering problem and propose a non-intrusive solution called Prism. Prism adopts a coarse-to-fine clustering strategy. It first partitions instances into coarse-grained chunks based on communication patterns. Within each chunk, Prism further groups instances with similar resource usage patterns to produce fine-grained functional clusters. Such a design reduces noises in the data and allows Prism to process massive instances efficiently. We evaluate Prism on two datasets collected from the real-world production environment of Huawei Cloud. Our experiments show that Prism achieves a v-measure of ~0.95, surpassing existing state-of-the-art solutions. Additionally, we illustrate the integration of Prism within monitoring systems for enhanced cloud reliability through two real-world use cases.

cs.SE↗

arXiv2023

Alioth: A Machine Learning Based Interference-Aware Performance Monitor for Multi-Tenancy Applications in Public Cloud

Multi-tenancy in public clouds may lead to co-location interference on shared resources, which possibly results in performance degradation of cloud applications. Cloud providers want to know when such events happen and how serious the degradation is, to perform interference-aware migrations and alleviate the problem. However, virtual machines (VM) in Infrastructure-as-a-Service public clouds are black-boxes to providers, where application-level performance information cannot be acquired. This makes performance monitoring intensely challenging as cloud providers can only rely on low-level metrics such as CPU usage and hardware counters. We propose a novel machine learning framework, Alioth, to monitor the performance degradation of cloud applications. To feed the data-hungry models, we first elaborate interference generators and conduct comprehensive co-location experiments on a testbed to build Alioth-dataset which reflects the complexity and dynamicity in real-world scenarios. Then we construct Alioth by (1) augmenting features via recovering low-level metrics under no interference using denoising auto-encoders, (2) devising a transfer learning model based on domain adaptation neural network to make models generalize on test cases unseen in offline training, and (3) developing a SHAP explainer to automate feature selection and enhance model interpretability. Experiments show that Alioth achieves an average mean absolute error of 5.29% offline and 10.8% when testing on applications unseen in the training stage, outperforming the baseline methods. Alioth is also robust in signaling quality-of-service violation under dynamicity. Finally, we demonstrate a possible application of Alioth's interpretability, providing insights to benefit the decision-making of cloud operators. The dataset and code of Alioth have been released on GitHub.

cs.DC↗

arXiv2023

Audience Expansion for Multi-show Release Based on an Edge-prompted Heterogeneous Graph Network

In the user targeting and expanding of new shows on a video platform, the key point is how their embeddings are generated. It's supposed to be personalized from the perspective of both users and shows. Furthermore, the pursue of both instant (click) and long-time (view time) rewards, and the cold-start problem for new shows bring additional challenges. Such a problem is suitable for processing by heterogeneous graph models, because of the natural graph structure of data. But real-world networks usually have billions of nodes and various types of edges. Few existing methods focus on handling large-scale data and exploiting different types of edges, especially the latter. In this paper, we propose a two-stage audience expansion scheme based on an edge-prompted heterogeneous graph network which can take different double-sided interactions and features into account. In the offline stage, to construct the graph, user IDs and specific side information combinations of the shows are chosen to be the nodes, and click/co-click relations and view time are used to build the edges. Embeddings and clustered user groups are then calculated. When new shows arrive, their embeddings and subsequent matching users can be produced within a consistent space. In the online stage, posterior data including click/view users are employed as seeds to look for similar users. The results on the public datasets and our billion-scale data demonstrate the accuracy and efficiency of our approach.

cs.SI↗

arXiv2023

Distribution of $δ$-connected components of self-affine sponge of Lalley-Gatzouras type

Let $(E, ρ)$ be a metric space and let $h_E\left( δ\right)$ be the cardinality of the set of $δ$-connected components of $E$. In literature, in case of that $E$ is a self-conformal set satisfying the open set condition or $E$ is a self-affine Sierpiński sponge, necessary and sufficient condition is given for the validity of the relation $ h_E(δ)\asymp δ^{-\dim_B E}, \text{ when }δ\to 0. $ In this paper, we generalize the above result to self-affine sponges of Lalley-Gatzouras type; actually in this case, we show that there exists a Bernoulli measure $μ$ such that for any cylinder $R$, it holds that $ h_R(δ)\asymp μ(R) δ^{-\dim_B E}, \text{ when }δ\to 0. $

math.GT↗

arXiv2023

Heterogeneous Anomaly Detection for Software Systems via Attentive Multi-modal Learning

Prompt and accurate detection of system anomalies is essential to ensure the reliability of software systems. Unlike manual efforts that exploit all available run-time information, existing approaches usually leverage only a single type of monitoring data (often logs or metrics) or fail to make effective use of the joint information among multi-source data. Consequently, many false predictions occur. To better understand the manifestations of system anomalies, we conduct a comprehensive empirical study based on a large amount of heterogeneous data, i.e., logs and metrics. Our study demonstrates that system anomalies could manifest distinctly in different data types. Thus, integrating heterogeneous data can help recover the complete picture of a system's health status. In this context, we propose HADES, the first work to effectively identify system anomalies based on heterogeneous data. Our approach employs a hierarchical architecture to learn a global representation of the system status by fusing log semantics and metric patterns. It captures discriminative features and meaningful interactions from multi-modal data via a novel cross-modal attention module, enabling accurate system anomaly detection. We evaluate HADES extensively on large-scale simulated and industrial datasets. The experimental results present the superiority of HADES in detecting system anomalies on heterogeneous data. We release the code and the annotated dataset for reproducibility and future research.

cs.SE↗

arXiv2023

Heterogeneous Anomaly Detection for Software Systems via Semi-supervised Cross-modal Attention

Prompt and accurate detection of system anomalies is essential to ensure the reliability of software systems. Unlike manual efforts that exploit all available run-time information, existing approaches usually leverage only a single type of monitoring data (often logs or metrics) or fail to make effective use of the joint information among different types of data. Consequently, many false predictions occur. To better understand the manifestations of system anomalies, we conduct a systematical study on a large amount of heterogeneous data, i.e., logs and metrics. Our study demonstrates that logs and metrics can manifest system anomalies collaboratively and complementarily, and neither of them only is sufficient. Thus, integrating heterogeneous data can help recover the complete picture of a system's health status. In this context, we propose Hades, the first end-to-end semi-supervised approach to effectively identify system anomalies based on heterogeneous data. Our approach employs a hierarchical architecture to learn a global representation of the system status by fusing log semantics and metric patterns. It captures discriminative features and meaningful interactions from heterogeneous data via a cross-modal attention module, trained in a semi-supervised manner. We evaluate Hades extensively on large-scale simulated data and datasets from Huawei Cloud. The experimental results present the effectiveness of our model in detecting system anomalies. We also release the code and the annotated dataset for replication and future research.

cs.SE↗

arXiv2023

Physiologically-Based Pharmacokinetic Modeling of Blood Clearance of Liver Fluorescent Markers for the Assessment of the Degree of Hepatic Ischemia-Reperfusion Injury

During liver transplantation, ischemia-reperfusion injury (IRI) is inevitable and decreases the overall success of the surgery. While guidelines exist, there is no reliable way to quantitatively assess the degree of IRI present in the liver. Our recent study has shown a correlation between the bile-to-plasma ratio of FDA-approved sodium fluorescein (SF) and the degree of hepatic IRI, presumably due to IRI-induced decrease in the activity of the hepatic multidrug resistance-associated protein 2 (MRP2); however, the contribution of SF blood clearance via the bile is still convoluted with other factors, such as renal clearance. In this work, we sought to computationally model SF blood clearance via the bile. First, we converted extant SF fluorescence data from rat whole blood, plasma, and bile to concentrations using calibration curves. Next, based on these SF concentration data, we generated a liver-centric, physiologically-based pharmacokinetic (PBPK) model of SF liver uptake and clearance via the bile. Model simulations show that SF bile concentration is highly sensitive to a change in the activity of hepatic MPR2. These simulations suggest that SF bile clearance along with the PBPK model can be used to quantify the effect of IRI on the activity of MRP2.

q-bio.TO↗

arXiv2022

Managing Service Dependency for Cloud Reliability: The Industrial Practice

Interactions between cloud services result in service dependencies. Evaluating and managing the cascading impacts caused by service dependencies is critical to the reliability of cloud systems. This paper summarizes the dependency types in cloud systems and demonstrates the design of the Dependency Management System (DMS), a platform for managing the service dependencies in the production cloud system. DMS features full-lifecycle support for service reliability (i.e., initial service deployment, service upgrade, proactive architectural optimization, and reactive failure mitigation) and refined characterization of the intensity of dependencies.

cs.DC↗

arXiv2022

Adaptive Performance Anomaly Detection for Online Service Systems via Pattern Sketching

To ensure the performance of online service systems, their status is closely monitored with various software and system metrics. Performance anomalies represent the performance degradation issues (e.g., slow response) of the service systems. When performing anomaly detection over the metrics, existing methods often lack the merit of interpretability, which is vital for engineers and analysts to take remediation actions. Moreover, they are unable to effectively accommodate the ever-changing services in an online fashion. To address these limitations, in this paper, we propose ADSketch, an interpretable and adaptive performance anomaly detection approach based on pattern sketching. ADSketch achieves interpretability by identifying groups of anomalous metric patterns, which represent particular types of performance issues. The underlying issues can then be immediately recognized if similar patterns emerge again. In addition, an adaptive learning algorithm is designed to embrace unprecedented patterns induced by service updates or user behavior changes. The proposed approach is evaluated with public data as well as industrial data collected from a representative online service system in Huawei Cloud. The experimental results show that ADSketch outperforms state-of-the-art approaches by a significant margin, and demonstrate the effectiveness of the online algorithm in new pattern discovery. Furthermore, our approach has been successfully deployed in industrial practice.

cs.SE↗